Open CYJnrclw opened 3 days ago
The data processing part of the project was failing due to a mismatch between the column names in the dataset and the expected feature names in the code. Specifically, the CSV file used MTML_*
column names, while the code expected PSML_*
feature names.
I have resolved this issue in two ways:
Code Modification: Updated the datagen.py
script to automatically map MTML_*
columns to PSML_*
, ensuring that the existing dataset can be used without modification.
CSV Column Rename: Alternatively, I created a modified version of the CSV file (Suturing_S02_T01_renamed.csv
) that renames all MTML_*
columns to PSML_*
, matching the expected format in the code.
Use the Code Modification:
datagen.py
that transforms the MTML_*
feature names to PSML_*
. You can integrate this function to handle datasets with different column name prefixes.Use the Renamed CSV:
Let me know which approach works best for your setup! Also I will be glad to open a pull request if you would like me to.
@TBJr Thanks a lot for suggesting this. We understand that our data preprocessing code expects all the raw data to be in a very specific format, which is not always practical. I also made another branch called "gesture" which has some more information and sample structure of the csv after preprocessing.
Thanks a lot for this change, and please feel free to create a PR. I will merge it. Thanks!
thanks
Why can't the data processing part of this code run?